JP2010191670A - Storage system, capacity management method and management computer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a storage system for making volume capacity proper among file management computers sharing a storage device. <P>SOLUTION: In this storage system, a capacity management computer 120 acquires statistical information, such as, access history from the file management computers 110 and the storage device 180. The capacity management computer 120 executes capacity optimization processing for the storage device, according to a designated date. The capacity management computer 120 calculates optimum capacity for each file system, based on the statistical information by reallocating free capacity of volumes according to a prescribed standard (e.g. a capacity consuming speed); sends reduction instruction of the file system to the file management computer 110b that manages the file system of a capacity reduction target; sends a volume capacity change instruction to the storage device 180 having the volume; and sends an expansion instruction of the file system to the file management computer 110a managing the file system of a capacity increase target according to expanded volume capacity. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a storage system, a capacity management method, and a management computer that can optimize a volume capacity among file management computers that share a storage device.

  As a basic function in a storage device, an automatic volume expansion technique is disclosed. For example, according to Patent Document 1, in a volume pool including a plurality of virtual volumes, when a virtual volume is created, physical blocks are not allocated to blocks constituting the virtual volume, and an actual use request (such as a write I / O request) is issued. When issued, a physical block is allocated from the volume pool to the target block of the request. By using automatic volume expansion technology to ensure that the virtual volume capacity is sufficiently large and the volume pool capacity is moderate, the administrator basically only monitors the availability of the volume pool. The virtual volume capacity expansion work is freed.

  Also disclosed is a technique for changing a storage volume in accordance with the characteristics of a file. For example, according to Patent Document 2, the storage device classifies the volumes into storage tiers called storage classes according to the characteristics of the disks constituting the volumes. In the invention of Patent Document 2, in a file system constructed on a volume, a file can be migrated between different storage classes according to dynamic characteristics and static attributes. In addition, by making it possible to specify a file with the same file path after migration as to the host that references the file system via the network, it is possible to conceal changes in the file position due to migration from the host. it can.

  Patent Document 3 discloses a technique capable of concealing a change in file position due to migration from a host as in Patent Document 2. While Patent Document 2 conceals file movement between volume and file systems having different characteristics, Patent Document 3 differs in that file movement between a plurality of computers managing the file system is concealed.

JP 2003-15915 A JP 2006-216070 A JP 2006-164111 A

  In managing the storage device, the performance and capacity of the storage device are the main management targets. In general, the performance and capacity of a storage device can be improved by adding a storage medium such as an HDD (Hard Disk Drive), a tape, or a flash memory included in the storage device. However, the addition of a storage medium leads to an increase in management cost such as installation area, power consumption, and MTTF (Mean Time To Failure) of the entire storage device as well as equipment introduction cost. On the other hand, in general, the entire capacity and performance of the storage device are rarely used. For example, it is said that the capacity usage rate of a large storage device in a company is about 35%.

  Because of these factors, for example, when there is a shortage of free space in some of the multiple volumes provided by the storage device, the storage medium is used due to insufficient performance and space even though other volumes have free space. It may be necessary to add.

  A similar argument holds true for a file system built on a volume provided by a storage device. In a general file system, a file system having the same size as the volume capacity is created when the file system is created, and file data, file management information, and information for managing the storage device usage status are stored in the volume. . Therefore, when the file system stores only a file smaller than the entire capacity of the file system, the file system has a free capacity, but the entire capacity of the volume is used and there is no free capacity. Therefore, when the performance or capacity of a certain file system is insufficient, or even when another file system has free capacity, it is necessary to add a storage medium.

  Here, if there is free space in a certain file system, if the free space can be used from another file system, the capacity shortage can be solved without adding a storage medium, and the usage rate of the storage device is increased. be able to.

  For this purpose, as one solution, there is a method of reducing the volume usage by reducing the operating file system having free capacity. Then, by assigning a volume capacity that is no longer used by the file system to a file system having no free capacity, it is possible to solve the shortage of free capacity. However, the reduction of the file system generally involves rearranging the data in the volume, which is a process that places a heavy load on the storage device and should not be performed frequently.

  An object of the present invention is to improve the usage rate of a storage device by distributing free capacity while reducing the frequency of file system reduction and reducing the overhead of performance. The overhead refers to a process that is indirectly and additionally required when the process proceeds, and a load generated by the process.

  The present invention further solves another problem. Even in a state other than the capacity shortage, the volume capacity used by another file system may be required. For example, when there are two file systems that use high-speed volumes and multiple file systems that use low-speed volumes, two computers use the file systems on the high-speed volumes one by one. Assume the situation.

  If one computer A needs more high-speed volumes due to performance problems, it is necessary to have the free capacity of the high-speed volume of the other computer B divided. If there is no free space in the file system, the file system cannot be reduced. Therefore, if the computer B uses the full capacity of the file system of a high-speed volume, even if the computer B does not require so much performance. The capacity of the high-speed volume cannot be divided into the computer A. Such a situation occurs not only in performance but also in performing capacity adjustment between volumes and file systems having different characteristics such as reliability and function.

  An object of the present invention is to provide a partial capacity of a volume used by the file system to another file system having the same characteristics even when the file system having the specific characteristics is already used. To do.

  In order to solve the above-described problem, a storage system management computer (for example, the capacity management computer 120) collects file system performance and functions and past capacity change history for a file system group sharing a storage device. The management computer predicts the time until the free capacity of each file system runs out based on the collected information, and calculates the capacity of the file system that increases the time when the free capacity shortage occurs. Then, the management computer reduces and expands each file system.

  In addition, in a configuration where there are multiple volume / file systems with different characteristics, the management computer uses the information collected as described above to calculate the capacity of the file system for each characteristic so that the characteristics of the entire system can be utilized. Calculate the capacity of the file system and determine the appropriate capacity.

  When a certain file system requires a capacity smaller than the current capacity, the file in the file system is moved to a file having a different characteristic, and the free capacity of the file system is increased. Then, the management computer reduces the file system whose free capacity is increased, and allocates the unused volume capacity to another file system, thereby optimizing the entire system.

  According to the present invention, in a shared file system group of storage devices, it is possible to optimize the capacity configuration of the file system of the entire system while reducing the load accompanying the capacity change of the file system.

  Further, when the storage device has a volume / file system having different characteristics in the above environment, it is possible to perform capacity distribution between the volume / file system so that the effect of good characteristics can be optimized in the entire system.

It is a block diagram which shows the structural example of the computer system containing the storage system in 1st Embodiment. It is a block diagram which shows the detailed structural example of a file management computer. It is a block diagram which shows the detailed structural example of a memory | storage device. It is a block diagram which shows the detailed structural example of a capacity management computer. It is a block diagram which shows the example of 1 operation form of the storage system in 1st Embodiment. It is explanatory drawing which shows the example of the access information table in 1st Embodiment. It is a flowchart which shows an access information acquisition process. It is a flowchart which shows the capacity | capacitance optimization process in 1st Embodiment. 3 is a block diagram illustrating an example of an operation mode of a storage system after application of capacity optimization processing in the first embodiment. FIG. It is a block diagram which shows the example of 1 operation form of the storage system in 2nd Embodiment. It is explanatory drawing which shows the example of the access information table in 2nd Embodiment. It is a flowchart which shows the capacity | capacitance optimization process in 2nd Embodiment. It is a block diagram which shows the example of 1 operation | movement form of the storage system after the application of the capacity optimization process in 2nd Embodiment. It is a block diagram which shows the structural example of the computer system containing the storage system in 2nd Embodiment. It is explanatory drawing which shows the example of the table group which the file location management program in 2nd Embodiment stores. It is a block diagram which shows the structural example of the computer system containing the other storage system in 2nd Embodiment.

  Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited thereby.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described.
FIG. 1 is a block diagram illustrating a configuration example of a computer system including a storage system according to the first embodiment. In FIG. 1, the storage system 100a (100) includes file management computers 110a (110) and 110b (110), a storage device 180, a capacity management computer 120, wiring 160, and a management network 130 which is a management network. A file storage function is provided to the host computers 150a (150) and 150b (150) connected to the data networks 140a (140) and 140b (140). The file management computers 110a and 110b are computers provided with file systems 520a (520) and 520b (520), respectively, to which the present invention is applied. The storage device 180 has a storage medium and provides a storage area to the file management computers 110a and 110b. The file system 520 will be described later with reference to FIG.

  The file management computers 110a and 110b are connected to the storage device 180 via a wiring 160 and share the storage device 180. The wiring 160 does not need to be separate between the file management computers 110a and 110b as shown in FIG. 1, and may be shared. For example, when connecting by FC (Fibre Channel), the wiring 160 coming out of the storage device 180 can be shared by the file management computers 110a and 110b by sandwiching the FC-SW (Switch) therebetween. In addition, the wiring 160 can use a communication format such as iSCSI (Internet Small Computer System Interface) or FCoE (Fibre Channel on Ethernet (registered trademark)). The file management computers 110a and 110b, the capacity management computer 120, and the storage device 180 are connected to the management network 130, and use the management network 130 as a communication path for control information and file data.

  The file management computers 110 a and 110 b can manage the storage device 180 by sending control information to the storage device 180 via the management network 130 or the wiring 160.

  The capacity management computer 120 can manage the storage device 180 by sending control information to the storage device 180 via the management network 130. Further, the capacity management computer 120 can indirectly manage the storage device 180 by controlling the file management computers 110a and 110b via the management network 130.

  The file management computers 110a and 110b can create file systems 520a and 520b on the volumes 540a (540) and 540b (540) provided from the storage device 180, respectively. The user can perform operations such as file creation, deletion, and reading / writing via an input / output device connected to the file management computers 110a and 110b. The file management computers 110a and 110b are connected to the data networks 140a and 140b, respectively. The host computers 150a and 150b perform operations such as file creation, deletion, and reading / writing via the data networks 140a and 140b. be able to.

  For example, a protocol such as TCP / IP (Transmission Control Protocol / Internet Protocol) or NetBEUI (NetBIOS Extended User Interface) can be used for communication between computers. Moreover, NFS (Network File System), CIFS (Common Internet File System), etc. can be used for the protocol which performs a file access request on these communication protocols.

  The capacity management computer 120 is a computer that communicates with the file management computers 110a and 110b and the storage device 180 via the management network 130 and performs processing of the present invention such as reception of statistical information and transmission of management information. The capacity management computer 120 may be connected to the data networks 140a and 140b and receive statistical information from other devices connected to the data networks 140a and 140b.

  FIG. 2 is a block diagram illustrating a detailed configuration example of the file management computer. The file management computer 110 has a control device 230, a memory 250, a network interface 210 (network I / F), and a storage interface 240 (storage I / F). Furthermore, the file management computer 110 can also have an input / output interface 220 (input / output I / F).

  The file management computer 110 communicates with the storage device 180 via the storage interface 240, and performs input / output processing to the volume provided by the storage device 180 and control of the storage device 180.

  The file management computer 110 can communicate with a computer connected to the management network 130 and the data networks 140a and 140b through a network interface 210 using a protocol such as TCP / IP or NetBEUI. In addition, a file access request by NFS, CIFS, or the like can be received.

  The control device 230 controls the file management computer 110 by operating according to a program stored in the memory 250. The memory 250 has a file system program 260, an access monitoring program 270, a storage device management program 280, a management support program 290, a file server program 295, and an access history information table 275.

  The file system program 260 is a program that stores information in a file format on a volume provided by the storage device 180 and provides a function that allows a user of the file management computer 110 to operate data on the volume in a file format. The file system program 260 manages each file with a unique name composed of a numerical value, a character string, and a symbol string. The user of the file management computer 110 can specify the operation target file by passing the file name to the file management computer 110. The access monitoring program 270 is a program that monitors the operation of the file system program 260, information processed by the storage interface 240 and the network interface 210, and stores statistical information such as the amount of data in the access history information table 275.

  The storage device management program 280 is a program that sends instructions such as volume addition / deletion / capacity change to the storage device 180. The management support program 290 executes the access monitoring program 270 and the storage device management program 280 in response to a management request sent from another computer via the network interface 210, and stores the contents of the access history information table 275. It is a program that returns.

  Further, the memory 250 can have a file server program 295. The file server program 295 is a program that calls the file system program 260 and performs file operations in response to a file operation request sent by a protocol such as NFS or CIFS via the network interface 210.

  A keyboard, mouse, display, printer, etc. are connected to the input / output interface 220, and the file management computer 110 user manages the file management computer 110 or the file on the file system managed on the file management computer 110. Can be operated.

  FIG. 3 is a block diagram illustrating a detailed configuration example of the storage device. The storage device 180 includes a control device 320, a storage interface 310 (storage I / F), a memory 330, and storage media 340a (340), 340b (340), and 340c (340). Further, the storage device 180 may include a network interface 350 (network I / F).

  The control device 320 extracts a part of the capacity from a combination of one or a plurality of storage media 340, forms a volume, and connects the volume ID and capacity to a computer connected by the wiring 160 via the storage interface 310. Notify information. Therefore, a computer connected by the wiring 160 via the storage interface 310, when inputting / outputting data to / from the storage device 180, uniquely specifies the data storage position by specifying the volume ID and the position in the volume. Can do. The control device 320 can also handle each of the storage media 340a to 340c as one volume. In addition, when the control device 320 receives control information from a computer connected by the wiring 160 via the storage interface 310, the control device 320 controls each element in the storage device 180 according to the contents.

  The storage medium 340 includes a magnetic disk such as an HDD, an optical disk such as a DVD (Digital Versatile Disc), a tape device, a flash memory, an SSD (Solid State Drive), or a disk array device including a plurality of HDDs and control devices. A device capable of storing information can be used. Further, by using a storage device 180 different from the storage device 180 included in the storage medium 340, the plurality of storage devices 180 may form a hierarchical structure. The storage media 340 need not all be the same medium, and media having different speed, capacity, and characteristics can be combined.

  The storage device 180 can use a part of the memory 330 as the disk cache 380. The memory 330 can use the disk cache 380 as a cache area for prefetching information stored in the storage medium 340 and delaying writing to the storage medium 340.

  Further, the storage device 180 can have a statistical information table 370 inside the memory 330. The statistical information table 370 can store statistical information such as volume usage information, cache memory usage, and processed input / output volume in the storage device 180.

  When the storage device 180 includes the network interface 350, control information of the storage device 180 and the contents of the statistical information table 370 can be transmitted and received from a computer connected via the network interface 350 via the management network 130.

  FIG. 4 is a block diagram illustrating a detailed configuration example of the capacity management computer. The capacity management computer 120 has a control device 430, a memory 450, and a network interface 410 (network I / F). Furthermore, the capacity management computer 120 can also have an input / output interface 420 (input / output I / F).

  The capacity management computer 120 is connected to the management network 130 via the network interface 410. By using this, the user of the capacity management computer 120 can operate the capacity management computer 120 from another computer connected to the management network 130. A user of the capacity management computer 120 can also operate the capacity management computer 120 via the input / output interface 420.

  The control device 430 controls the capacity management computer 120 by operating according to a program stored in the memory 450. The memory 450 has an access information acquisition program 470 (statistical information acquisition program), a capacity optimization program 480 (allocated capacity calculation program), and an access information table 475.

  The access information acquisition program 470 acquires the access history information table 275 possessed by the file management computer 110 and the statistical information table 370 possessed by the storage device 180 via the network interface 410, processes them, and stores them in the access information table 475. It is. The access information acquisition program 470 can collect statistical information from computers connected to the management network 130 and store it in the access information table 475 other than the file management computer 110 and the storage device 180. Specifically, the access information table 475 includes a file system statistical information table 600 (see FIG. 6A) and a storage device statistical information table 650 (see FIG. 6B) which will be described later.

  The capacity optimization program 480 refers to the access information table 475, calculates the optimum capacity in the volume that the storage device 180 allocates to each file management computer 110, and further, based on the calculation result, stores the file management computer 110 and storage This is a program for instructing the apparatus 180 to change the capacity.

  In FIG. 1 and FIG. 4, the capacity management computer 120 is described as one independent computer. However, as long as it has each element constituting the capacity management computer 120, the file management computer 110 and storage The device 180 can also operate as the capacity management computer 120. For example, the file management computer 110 has a control device 230, a memory 250, a network interface 210, and an input / output interface 220. In addition to the memory 250, an access information acquisition program 470, an access information table 475, a capacity optimization program 480, a system management By having the program 490, it can also operate as the capacity management computer 120.

  FIG. 5 is a block diagram illustrating an example of an operation mode of the storage system according to the first embodiment. In the storage system 100a (see FIG. 1), the file management computer 110a having the ID "NAS1" and the file management computer 110b having the ID "NAS2" share one storage device 180 having the ID "STR". is doing. The storage device 180 has a storage medium group 550 having an ID of “RG1”, and creates two volumes 540a and 540b having names “Vol1” and “Vol2” therefrom. In this example, the storage medium group 550 has a capacity of 1000 GB, and the volumes 540a and 540b have a capacity of 500 GB by half. However, it is assumed that the file system 520a has more file access requests than the file system 520b and consumes a large amount of capacity.

  The file management computer 110a creates a file system 520a having an ID of “FS1” on the volume 540a. The file system 520a already uses all the capacity for the total capacity of 500 GB. Therefore, the file system 520a cannot store any more files and is in a state where continuous operation is not possible. Note that 500 GB / 500 GB represents consumed capacity / allocated capacity, and means that 500 GB is consumed (used) with respect to the allocated capacity 500 GB.

  The file management computer 110b creates a file system 520b having an ID of “FS2” on the volume 540b. The file system 520b already uses 200 GB for the total capacity of 500 GB.

  FIG. 6 is an explanatory diagram illustrating an example of an access information table. Reference is made to FIGS. 1 and 4 as appropriate. The access information table 475 of the capacity management computer 120 includes a file system statistical information table 600 shown in FIG. 6A and a storage device statistical information table 650 shown in FIG. 6B.

  The file system statistical information table 600 is a kind of an access information table 475 that is generated by the access information acquisition program 470 of the capacity management computer 120 by counting the access history information tables 275 of the file management computers 110a and 110b. The file system statistical information table 600 stores one line of data 620 and 621 for each file system in the storage system 100a. In addition to the file system ID 612 (file system identification ID), the file system statistical information table 600 includes a node ID 611 for identifying the file management computer having the file system, and a volume ID 613 (volume volume) indicating the volume in which the file system is created. Identification ID), capacity changeable 614 (flag) indicating whether the file system capacity can be changed, total file system capacity 615 (volume allocated capacity), and free capacity 616 indicating an unused area of the file system. The file system statistical information table 600 stores the capacity consumption speed 617 in each file system.

  Specifically, from the data 620, the file system “FS1” (520a) uses the volume “Vol1” (540a), and there is no free capacity (0 GB) with respect to the total capacity 500 GB, and the capacity consumption speed It can be seen that is 200 MB / day. In addition, from the data 621, the file system “FS2” (520b) uses the volume “Vol2” (540b) and has a free capacity of 300 GB with respect to the total capacity of 500 GB, and the capacity consumption speed is 100 MB / day. I know that there is.

  The information in the file system statistical information table 600 listed here is an example, and other information can be stored. For example, detailed information for each file access frequency and date and time of capacity consumption can be stored.

  The storage device statistical information table 650 is a type of the access information table 475 generated by the access information acquisition program 470 of the capacity management computer 120 by adding up the statistical information table 370 of the storage device 180. The storage device statistical information table 650 stores one row of data 670 and 671 for each volume in the storage system 100a. The storage device statistical information table 650 stores storage device ID 661, storage medium ID 662, maximum storage medium performance 663, volume ID 664, volume capacity 665, and volume assignment destination node ID 666 as statistical information regarding individual volumes.

  Specifically, from the data 670, the storage medium group “RG1” (550) of the storage device “STR” (180) has a maximum performance of 300 MB / s, and the volume “Vol1” (540a) includes a file. It can be seen that a capacity of 500 GB is allocated to the management computer “NAS1” (110a), and a capacity of 500 GB is allocated to the volume “Vol2” (540b) to the file management computer “NAS2” (110b).

  The information in the storage device statistical information table 650 given here is an example, and other information can be stored. For example, detailed information for each date and time of volume response time, power consumption, and input / output amount can be stored.

  Further, the access information table 475 is not limited to the file system statistical information table 600 and the storage device statistical information table 650, and can store statistical information related to the storage system 100a. For example, the usage rate of the control device 230 in the file management computers 110a and 110b, the bandwidth usage rate of the wiring 160, the usage status of the disk cache 380 in the storage device 180, and the like may be stored.

Next, processing will be described.
(Access information acquisition process)
FIG. 7 is a flowchart showing access information acquisition processing. Access information acquisition processing S700 by the access information acquisition program 470 is processing for acquiring access information from the file management computer 110, the storage device 180, and an external device and storing it in the access information table 475.

  The capacity management computer 120 receives a specific event from a device connected to the management network 130, such as when a start instruction is issued by the administrator, a preset periodic cycle, or a lack of free space in the file system. When notified, the access information acquisition process S700 is executed.

  In step S710, the capacity management computer 120 sends an acquisition request for the access history information table 275 to one or more file management computers 110 connected to the management network 130, and the access history information returned by the file management computer 110 in response. Only the items referred to by the capacity optimization program 480 are extracted from the table 275 to generate the file system statistical information table 600.

  In step S720, the capacity management computer 120 sends a statistical information table 370 acquisition request to one or more storage devices 180 connected to the management network 130, and among the statistical information tables 370 returned by the storage device 180 in response, Only the items referred to by the capacity optimization program 480 are extracted to generate the storage device statistical information table 650.

  If there is a device connected to the management network 130 other than the file management computer 110 and the storage device 180 in step S730, the capacity management computer 120 obtains statistical information on the device having the information required by the capacity optimization program 480. You can make a request. For example, the network bandwidth usage rate can be acquired from the network bandwidth monitoring device.

  Steps S710, S720, and S730 are independent processes, and are not necessarily performed continuously, and may be performed concurrently. Further, step S710, step S720, and step S730 are not necessarily performed at the same timing. For example, a procedure may be taken in which step S710 is performed every 10 minutes, step S720 is performed every 20 minutes, and step S730 is performed at the time of an administrator instruction.

(Capacity optimization processing)
FIG. 8 is a flowchart showing the capacity optimization process in the first embodiment. The capacity optimization process S800 is a process by the capacity optimization program 480. The capacity management computer 120 receives a specific event from a device connected to the management network 130, such as when a start instruction is issued by the administrator, a preset periodic cycle, or a lack of free space in the file system. The capacity optimization process S800 is executed when notified.

  In step S810, the capacity management computer 120 first calculates the optimum capacity of each volume allocated to each file management computer and file system using the access information table 475.

  The optimum capacity (appropriate capacity) described here is the capacity configuration for each volume that raises the standard to the maximum source in the entire storage system in the specific standard expected by the administrator such as performance, storage usage rate, reliability, power consumption etc. Indicates. The optimum capacity can be obtained from the standard instructed by the administrator and statistical information such as the current total capacity 615, free capacity 616, capacity consumption speed 617 and the like shown in FIG.

  In the calculation process of the optimum capacity, the total volume capacity is not changed between volumes that have file systems that can change the capacity using the same storage device and storage medium, and the usage of each file system is less than the amount used. The capacity can be moved within a range. However, if the storage medium capacity can be added by the administrator, the total volume capacity may be increased.

  As a specific example of step S810, an example of the optimum capacity calculation is shown based on the statistical information (file system statistical information table 600, storage device statistical information table 650) in FIG. It is assumed that the administrator of the storage system instructs the optimum capacity calculation that maximizes the usage rate of the storage system, that is, the longest time until the shortage of free capacity of each volume occurs. In the example of FIG. 6, the two volumes “Vol1” (540a) and “Vol2” (540b) in which the two file systems “FS1” and “FS2” are constructed belong to the storage medium group “RG1” (550). The capacity can be moved.

  The capacity management computer 120 calculates the total value of the free capacity 616 in the file system group using the same storage medium, and distributes the total free capacity at the same ratio as the ratio of the capacity consumption speed 617 of each file system. In order to secure the free space 616 after distribution, the total capacity 615 of each file system is set to the used capacity of the file system, that is, the difference between the total capacity 500 before calculation and the free capacity 616, Change to the added value.

  In FIG. 6, since the total free capacity of the file system ID “FS1” of the data 620 and the file system ID “FS2” of the data 621 is 300 GB, the capacity management computer 120 uses this total free capacity as a ratio of the capacity consumption speed. 200 MB / day: Allotted at the same rate as 100 MB / day, the total capacity of the file system “FS1” (520a) is 700 GB (the combined capacity of the used amount 500 GB and the free capacity 200 GB), and the file system “FS2”. "The total capacity of (520b) is calculated as the optimum capacity by 300 GB (the combined capacity of the used amount 200 GB and the free capacity 100 GB) at which the free capacity becomes 100 GB.

  Returning to FIG. 8, in the subsequent step S820, the capacity management computer 120 owns the corresponding file system for each file system having a volume for which the volume capacity smaller than the current volume capacity is calculated to be optimal in step S810. A file system reduction instruction request is sent to the file management computer 110.

  When the file management computer 110 receives the file system reduction instruction request, the file system program 260 operates and the file system reduction processing is performed.

  An example of the reduction process of the first file system is shown below. In order to reduce the file system, the file management computer 110 makes unused the volume area after the reduced volume capacity. If the file data or management data is already placed in the area after the reduced volume capacity in order to make the volume area unused, the data is moved to fit within the volume capacity after the reduction. To do. Further, inode management information and free area management information, which are file system management information, are reduced so that a volume area after the reduced volume capacity is not managed. As a result, the volume area after the reduced volume capacity can be unused.

  Further, as the second file system reduction process, when the volume assigned to the file management computer 110 is a virtual volume in the storage device 180 according to the invention shown in Patent Document 1, the file management computer 110 is a storage device. When 180 physical block allocation mechanisms can be controlled, the next file system reduction process can be performed. First, the file management computer 110 calculates the number of physical blocks that need to be deallocated before reaching the instructed volume capacity. If there is an unused physical block in the file system, the file management computer 110 issues a physical block allocation cancellation instruction to the storage device as many times as necessary. Even if all unused physical blocks are deallocated, if the specified volume capacity is exceeded, the file management computer 110 selects a plurality of partially used physical blocks, and the file system within those physical blocks. The data is moved while maintaining consistency, and is divided into physical blocks in which the entire area is used and physical blocks that are not used at all. Then, an instruction to deallocate a physical block that is not used at all is issued.

  In the reduction process of the second file system, data movement occurs only in a partially used physical block. Therefore, in a file system that consumes the entire area in the volume evenly, The file system can be reduced in a shorter time than the reduction process.

  Returning to FIG. 8, in step S830, the capacity management computer 120 changes the volume capacity to the storage device 180 that owns the volume for each volume for which the volume capacity different from the current volume capacity is optimal in step S810. Send instructions.

  In the subsequent step S840, the capacity management computer 120 sends to the file management computer 110 that owns the corresponding file system for each file system that has a volume for which the volume capacity larger than the current volume capacity is optimal in step S810. Send a file system expansion instruction request.

  When the file management computer 110 receives the file system enlargement instruction request, the file system program 260 operates to perform the file system enlargement process.

  As an example of file system expansion processing, the file management computer 110 expands inode management information and free space management information, which are file system management information, for newly expanding the file system. Can be managed. As a result, the file management computer 110 can use the area added by the volume expansion, and the expansion of the file system is completed.

  Returning to FIG. 8, the capacity management computer 120 can realize capacity optimization of the entire storage system by completing a series of processing up to step S840.

  FIG. 9 is a block diagram illustrating an example of an operation mode of the storage system after application of the capacity optimization process in the first embodiment. FIG. 9 shows a state after the capacity optimization process S800 is applied to the storage system 100a shown in FIG. The storage medium group 550 included in the storage device 180 has a capacity of 1000 GB without changing before and after processing. However, the volumes 540a and 540b generated from the storage medium group 550 were both 500 GB before the capacity optimization process S800 was applied. However, after the capacity optimization process S800 was applied, the volume 540a was 700 GB and the volume 540b was It is 300GB. As the volume capacity is changed, the maximum capacity of the file systems 520a and 520b is also changed. As a result, without adding the storage medium group 550, the free capacity is added to the file system 520a that is in a capacity shortage state, and continuous operation is possible. Further, after the application of the capacity optimization process S800, according to the file system statistical information table 600, when the file systems 520a and 520b use capacity at the current capacity consumption speed 617, capacity shortage occurs simultaneously after 1000 days. This is a state in which the entire capacity of the storage medium group 550 has been used up, and the time until the shortage of free capacity surely has been maximized.

(Second Embodiment)
Hereinafter, a second embodiment of the present invention will be described.
The second embodiment can be applied to a storage system having a mechanism capable of concealing a name change from an external computer when moving a file between file systems and inputting / outputting a file using the same file name before and after the movement. . Regarding the concealment mechanism for the name change, there are a method using Patent Document 2 and a method using Patent Document 3. As the second embodiment, the respective methods will be described as a first name change concealment mechanism example (see FIGS. 14 and 15) and a second name change concealment mechanism example (see FIG. 16).

(Example of first name concealment mechanism)
FIG. 14 is a block diagram illustrating a configuration example of a computer system including a storage system according to the second embodiment. FIG. 14 shows a computer system including a storage system 100b (100) to which the technique of Patent Document 2 is applied. It is assumed that the memory 250 of the file management computers 110c (110) and 110d (110) included in the storage system 100b has a file location management program 285. The file location management program 285 has a file name management table 1600 shown in FIG.

  FIG. 15 is an explanatory diagram illustrating an example of a table group stored in the file location management program according to the second embodiment. The file name management table 1600 stores various pieces of meta information 1610, 1611, 1612, 1613, and 1614 such as file names and file handlers for each file managed by the file management computer 110 for each file. The file meta information has a reference to the file storage management table 1650.

  The file storage management table 1650 has information related to file storage such as a size 1651 and a block number 1652. Further, the file storage management table 1650 has a link destination node name 1661, a link destination file name 1662, and a link destination file handler 1663.

  When the file management computer 110 receives an input / output request for a file, the file management computer 110 refers to the file storage management table 1650 and is in a location described in the link destination node name 1661, the link destination file name 1662, and the link destination file handler 1663. I / O requests can be transferred to a file. When moving a file between file systems, by rewriting the link destination file information in the file storage management table, the file can be moved without changing the original file name, and the file can be moved from outside the file management computer 110. Can be concealed.

(Second example of renaming concealment mechanism)
FIG. 16 is a block diagram illustrating a configuration example of a computer system including other storage systems according to the second embodiment. FIG. 16 shows a computer system including a storage system 100c in which the technology of Patent Document 3 is applied to the storage system 100a. The storage system 100c is configured by adding an intermediate device 190 (accepting device) to the storage system 100a.

  In the method of Patent Document 3, when the host computers 150a and 150b in FIG. 1 input and output files stored in the file management computers 110c (110) and 110d (110), the host computers 150a and 150b are directly connected. Instead of communicating with the file management computers 110c and 110d, an input / output request is made to the intermediate device 190.

  The intermediate device 190 refers to the files in the file management computers 110c and 110d, and when stub information indicating a link to another file system is stored in the file contents, the file in which the link destination file is stored again An input / output request is sent to the management computers 110c and 110d. When the file moves between the file management computers 110c and 110d or within the computer, the intermediate device 190 creates the stub information and holds a link to the destination file with the original file name. Thereby, the intermediate device 190 can conceal the file movement between the file management computers 110c and 110d or within the computer.

  By using the first name change concealment mechanism or the second name change concealment mechanism, when the file management computers 110c and 110d have a plurality of file systems 1020a and 1020b and a plurality of file systems 1021a and 1021b, respectively, the host Since the computers 150a and 150b do not need to know in which file system the file that issued the processing request is stored, the storage system 100 uses the file system included in each of the file management computers 110c and 110d as an integrated file system. 1060 and 1061 can be shown to the host computers 150a and 150b.

  If the name change concealment mechanism is used, the storage system 100 can move the storage location of the file without changing the file name necessary for the host computers 150a and 150b to refer to the file. This process of moving a file between file systems while retaining the file name is called migration.

  FIG. 10 is a block diagram illustrating an example of an operation mode of the storage system according to the second embodiment. The operation mode shown in FIG. 10 is generally similar in configuration to the storage system 100a shown in FIG. 5, but differs in that the storage device 180 with the storage device ID “STR” has a plurality of storage medium groups 1050 and 1051.

  In the storage system 100 (100b or 100c), the file management computer 110c having the “NAS1” ID and the 110d having the file management computer ID “NAS2” share one storage device 180 having the “STR” ID. is doing.

  The storage device “STR” (180) has a storage medium group 1050 having an ID of “RG1”, and creates two volumes 1040a and 1040b having names “Vol1a” and “Vol1b” therefrom. In this example, the storage medium group 1050 has a capacity of 1000 GB, and the volumes 1040a and 1040b have a capacity of 500 GB by half. The storage device “STR” (180) has a storage medium group 1051 having an ID of “RG2”, and creates two volumes 1041a and 1041b having names “Vol2a” and “Vol2b” therefrom. . In this example, the storage medium group 1051 has a capacity of 4000 GB, and the volumes 1041a and 1041b have a capacity of 2000 GB by half.

  Here, it is assumed that the storage medium group 1050 and the storage medium group 1051 have different properties other than the capacity. For example, storage media groups having different input / output bandwidths, delay times, device reliability, prices, and the like are conceivable. Here, it is assumed that the storage medium group 1050 uses a high-speed storage medium such as an SSD, and the storage medium group 1051 uses a storage medium such as an HDD having a lower performance than the SSD, and the storage medium group 1050 inputs and outputs from the storage medium group 1051. An example will be described in which the bandwidth is high-performance.

  The file management computer 110c creates a file system 1020a having an ID of “FS1a” on the volume 1040a and a file system 1020b having an ID of “FS1b” on the volume 1041a. The file management computer 110d creates a file system 1021a having an ID of “FS2a” on the volume 1040b and a file system 1021b having an ID of “FS2b” on the volume 1041b. The file system 1020a and the file system 1021a have already used all the capacity with respect to the total capacity of 500 GB. The file system 1020b has a usage amount of 500 GB with respect to a total capacity of 2000 GB, and still has a free space of 1500 GB. The file system 1021b has a usage amount of 1000 GB with respect to a total capacity of 2000 GB, and still leaves 1000 GB of free space.

  Here, the file management computers 110c and 110d create an integrated file system in which the file system capacities are integrated by using the name concealment mechanism. The file management computer 110c integrates the file systems 1020a and 1020b to create an integrated file system “FS1c” (1060). The integrated file system 1060 combines the capacities of the file systems 1020a and 1020b, resulting in a total capacity of 2500 GB and a used capacity of 1000 GB.

  Similarly, the file management computer 110d integrates the file systems 1021a and 1021b to create an integrated file system “FS2c” (1061). The integrated file system 1061 has a total capacity of 2500 GB and a used capacity of 1500 GB by combining the capacities of the file systems 1021a and 1021b.

  FIG. 11 is an explanatory diagram illustrating an example of an access information table according to the second embodiment. The access information table 475 of the capacity management computer 120 includes a file system statistical information table 600A shown in FIG. 11A and a storage device statistical information table 650A shown in FIG.

  The file system statistical information table 600A stores one line of data 1120a, 1120b, 1121a, 1121b for each volume constituting the integrated file system. The file system statistical information table 600A has information similar to the file system statistical information table 600 (see FIG. 6A), and in addition to the file system ID 612, a node ID 611 for specifying a file management computer having the file system, a file A volume ID 613 indicating the volume in which the system is created, a capacity changeable 614 (flag) indicating whether the capacity of the file system can be changed, a total capacity 615 of the file system, and a free capacity 616 indicating an unused area of the file system. Have.

  The file system statistical information table 600A stores the current performance 1117 (access speed) and the required performance 1118 (request access speed) in each integrated file system. The current performance 1117 can be calculated from the past input / output speed with reference to the access history information table 275 of the file management computer 110. The required performance 1118 can be obtained by dividing the current performance 117 by the ratio that the host computer 150a, 150b could not respond to the file request because of waiting for input / output to the storage device 180. Alternatively, the required performance 1118 may be written directly into the file system statistical information table 600A by the storage system administrator via the input / output interface 420 or the network interface 410.

  The information in the file system statistical information table 600A mentioned here is an example, and other information can be stored. For example, detailed information for each file access frequency and date and time of capacity consumption can be stored.

  The storage device statistical information table 650A is a type of the access information table 475 generated by the access information acquisition program 470 of the capacity management computer 120 by adding up the statistical information table 370 of the storage device 180. The storage device statistical information table 650A stores one row of data 1170a, 1170b, 1171a, and 1171b for each volume in the storage system. Similar to the storage device statistical information table 650 (see FIG. 6B), the storage device statistical information table 650A includes a storage device ID 661, a storage medium ID 662, a maximum performance 663 of the storage medium, and a volume as statistical information regarding individual volumes. An ID 664, a volume capacity 665, and a volume assignment destination node ID 666 are stored. In addition, the storage device statistical information table 650 can also have other statistical information.

  For example, it is assumed that input / output requests to the integrated file system are distributed to each file system in accordance with the used capacity ratio of the file system of the integration source. Then, in this example, the input / output request to the integrated file system “FS1c” (1060) is the volume “Vol1a” (1040a) and the volume “Vol2a” (1041a) based on the used capacity ratio of the integration-source file systems 1020a and 1020b. To, the ratio is 1: 1. An input / output request to the integrated file system “FS2c” (1061) is sent from the integration source file systems 1021a and 1021b to the volume ID “Vol1b” (1040b) and the volume ID “Vol2b” (1041b). Sorted at a ratio of 1: 2.

  The total maximum performance of the storage medium group “RG1” (1050) and the storage medium group “RG2” (1051) is 600 MB / s from the maximum performance 663 of the storage device statistical information table 650A, and the integrated file systems 1060 and 1061 The total required performance is required performance 1118 to 450 MB / s of the file system statistical information table 600A. Therefore, the performance of the storage medium groups 1050 and 1051 sufficiently exceeds the required performance. However, considering the ratio of the input / output request distribution, the file system statistical information table 600A cannot satisfy the required performance 1118. For example, in the file system statistical information table 600A, in the integrated file system “FS1c” (1060), the current performance 1117 is 200 MB / s or less, and the required performance 1118 is 300 MB / s or less.

  FIG. 12 is a flowchart showing the capacity optimization process in the second embodiment. The capacity optimization process S1200 is a process by the capacity optimization program 480. The capacity management computer 120 receives a specific event from a device connected to the management network 130, such as when a start instruction is issued by the administrator, a preset periodic cycle, or a lack of free space in the file system. The capacity optimization process S1200 is executed when notified.

  In step S1210, the capacity management computer 120 first calculates the optimum capacity of each volume to be allocated to each file management computer and file system using the access information table 475. The optimum capacity described here indicates a capacity configuration for each volume that maximizes the standard of the entire storage system in a specific standard expected by the administrator such as performance, storage usage rate, reliability, and power usage. The optimum capacity can be obtained from the standard instructed by the administrator and statistical information such as the current total capacity 615, free capacity 616, capacity consumption speed 617, and the like.

  In the optimum capacity calculation step S1210 in the second embodiment, unlike the optimum capacity calculation step S810 (see FIG. 8) in the first embodiment, the file system is reduced even if the free capacity of each file system is insufficient. Calculate the optimal capacity as possible. However, the optimum capacity is calculated as long as the free capacity of the integrated file system in which the file systems are integrated is not short. In addition, in each file system in the integrated file system, the optimum capacity is calculated on the assumption that the used capacity is movable between each other.

  As a specific example of step S1210, an example of optimum capacity calculation based on the statistical information (file system statistical information table 600A, storage device statistical information table 650A) in FIG. 11 is shown. The administrator of the storage system 100 is instructed to calculate the optimum capacity that satisfies the required performance 1118 of each of the integrated file systems “FS1c” and “FS2c”.

An equation for the capacity and performance of each file system can be established from the following conditions (first condition to fifth condition).
As a first condition, the total use amount of the file system in each integrated file system does not change.
As a second condition, the capacity of each file system does not fall below the usage amount.
As a third condition, the total capacity of the file system including the volumes sharing the storage medium group does not exceed the capacity of the storage medium group.
As a fourth condition, the total required performance for a file system composed of volumes sharing a storage medium group does not exceed the performance of the storage medium group.
As a fifth condition, the required performance for the integrated file system is proportional to the usage ratio of each file system and becomes the required performance for each file system.

  In the volumes 1040a, 1041a, 1040b, and 1041b that store the file systems 1020a, 1020b, 1021a, and 1021b, the required capacities are T1a, T2a, T1b, and T2b, and the usage amounts are U1a, U2a, U1b, and U2b.

From the first condition,
U1a + U2a = 1000 GB,
U1b + U2b = 1500GB
The following equation is obtained.

From the second condition,
T1a ≧ U1a,
T2a ≧ U2a,
T1b ≧ U1b,
T2b ≧ U2b
The following equation is obtained.

From the third condition,
T1a + T1b ≦ 1000 GB,
T2a + T2b ≦ 4000GB
The following equation is obtained.

From the fourth and fifth conditions,
300 × U1a / (U1a + U2a) + 150 × U1b / (U1b + U2b) ≦ 400,
300 × U2a / (U1a + U2a) + 150 × U2b / (U1b + U2b) ≦ 200
The following equation is obtained.

  The total capacity of the integrated file system may or may not change before and after the calculation. For example, in this example, the required performance can be achieved only by changing the capacities of the volumes 1040a and 1040b, but in this case, the total capacities of the integrated file systems 1060 and 1061 change. If the administrator does not want to change the capacity of the integrated file system, the volume 1040a is enlarged by 300 GB, the volume 1040b is reduced by 300 GB, the volume 1041a is reduced by 300 GB, the volume 1041b is enlarged by 300 GB, and the entire integrated file is The system capacity can not be changed.

The optimal capacity can be obtained by calculating the capacity and the usage amount that satisfy the conditions by a method such as linear programming under these constraint conditions. In this example,
T1a = U1a = 800 GB,
T2a = 1700 GB, U2a = 200 GB,
T1b = U1b = 200 GB,
T2b = 2300GB, U2b = 1300GB
Satisfies these conditions.

  Returning to FIG. 12, in the subsequent step S1220, the capacity management computer 120 selects the corresponding file system for each file system in which the used capacity is reduced and the migration destination has free capacity based on the optimum capacity calculated in step S1210. A file migration instruction request is sent to the own file management computer 110.

  The file management computer 110 migrates the file in the file system that has received the migration instruction to another file system that is under the same integrated file system and has free space. The file management computer 110 uses the techniques of Patent Literature 2 and Patent Literature 3 to perform file migration without changing the file path seen from the host computers 150a and 150b, and move the file to the host computers 150a and 150b. Do not be conscious.

  For example, in the example of the optimum capacity calculation, the capacity management computer 120 uses the file system “FS2b” (1021a) for the file system “FS2b” because the file system “FS2b” (1021b) already has 1000 GB of free capacity. "(1021b) is instructed to migrate.

  Which file in the file system is to be migrated may be determined by the capacity management computer 120 or the file management computer 110. For example, it is possible to use a criterion such as placing a file for which input / output requests are frequently made in a file system using a high-speed storage medium group.

  In the subsequent step S1230, the capacity management computer 120 sends to the file management computer 110 that owns the corresponding file system for each file system having a volume whose volume capacity that is smaller than the current used capacity is optimal in step S1210. Send a file system reduction instruction request. The process in step S1230 is the same as step S820 in the first embodiment.

  In step S1240, the capacity management computer 120 sends a volume capacity change instruction to the storage device 180 that owns the volume for each volume for which the volume capacity different from the current volume capacity is calculated in step S1210. The process in step S1240 is the same as step S830 in the first embodiment.

  In the subsequent step S1250, the capacity management computer 120 sends the file management computer 110 that owns the corresponding file system to each file system having a volume for which the volume capacity larger than the current volume capacity is optimal in step S1210. Send a file system expansion instruction request. The processing in step S1250 is the same as step S840 in the first embodiment.

  In the subsequent step S1260, the capacity management computer 120 owns the corresponding file system for each file system whose used capacity has decreased based on the optimum capacity calculated in step S1210 and migration processing has not been performed in step S1220. A file migration instruction request is sent to the file management computer 110. When the file management computer 110 receives the migration instruction, the file management computer 110 performs the same processing as that instructed in step S1220.

  The capacity management computer 120 completes the series of processing up to step S1260, thereby realizing capacity optimization for improving the performance of the entire storage system.

  FIG. 13 is a block diagram illustrating an example of an operation mode of the storage system after application of the capacity optimization process in the second embodiment. FIG. 13 shows a state 1300 after the capacity optimization processing S1200 for improving performance is applied in the storage system 100 shown in FIG. As calculated by the capacity management computer 120 in step S1210, the file system 1020a constructed on the volume 1040a has a total capacity of 800 GB, of which the usage amount is 800 GB, and the file system 1021a constructed on the volume 1040b has a total capacity. 200 GB, of which the usage amount is 200 GB, and the file system 1020 b constructed on the volume 1041 a has a total capacity of 1700 GB, of which the usage amount is 200 GB, and the file system 1021 b constructed on the volume 1041 b has a total capacity of 2300 GB, Of these, the capacity can be optimized for a usage amount of 1300 GB.

  At this time, if the volume 1040a is 240 MB / s, the volume 1040b is 20 MB / s, the volume 1041a is 60 MB / s, and the volume 1041b is 130 MB / s, the maximum performance of the storage medium groups 1050 and 1051 will not be exceeded. In addition, the required performance of the integrated file systems 1060 and 1061 can be satisfied. As a result, continuous operation satisfying the required performance can be realized without adding the high-speed storage medium group 1050.

  The storage system 100 according to this embodiment includes a plurality of file management computers 110 that construct a file system on a volume of the storage device 180, and a capacity management computer 120 that manages and redistributes the allocated capacity allocated to each file system. With.

  The capacity management computer 120 collects, from the file management computer 110 and the storage device 180, a file system statistical information table 600 regarding the allocated capacity for each file system and a storage device statistical information table 650 regarding the volume capacity of the storage device 180. When the redistribution request for the allocated capacity of the file system is received, the total free capacity of the file system is obtained based on the file system statistical information table 600 and the storage device statistical information table 650 as predetermined performance information. An allocation capacity is calculated for each file system so as to satisfy, and a file system reduction instruction is transmitted to a file management computer that manages a file system whose allocation capacity is reduced among a plurality of file management computers, and Based on Send a volume capacity change instruction to the storage device transmits the file system enlargement instruction to the file management computer, wherein the allocated capacity of the plurality of file management computer manages the file system to expand.

  Here, in the case of the first embodiment, a file management computer that manages a file system whose allocated capacity is reduced among a plurality of file management computers corresponds to the file management computer 110b, and among the plurality of file management computers, A file management computer that manages a file system whose allocated capacity is expanded corresponds to the file management computer 110a. In the case of the second embodiment, as shown in FIGS. 10 and 13, the capacity of the file system (file system 1020a, 1020b or file system 1021a, 1021b) managed by the integrated file system 1060, 1061 is small. May increase or decrease. Therefore, a file management computer that manages a file system in which the allocated capacity is reduced among a plurality of file management computers and a file management computer that manages a file system in which the allocated capacity is expanded among a plurality of file management computers are both This corresponds to the file management computers 110c and 110d.

100 storage system 110 file management computer 120 capacity management computer (management computer)
130 Management Network 140 Data Network 150 Host Computer 160 Wiring 180 Storage Device 190 Intermediate Device (Reception Device)
230, 320, 430 Controller 250, 330, 450 Memory (storage unit)
260 File system program 270 Access monitoring program 275 Access history information table 280 Storage device management program 285 File location management program 290 Management support program 295 File server program 340 Storage medium 370 Statistical information table 380 Disk cache 470 Access information acquisition program (statistic information acquisition) program)
475 Access information table 480 Capacity optimization program (allocated capacity calculation program)
520a, 520b File system 540a, 540b Volume 550 Storage medium group 600 File system statistical information table (file system statistical information)
650 storage device statistical information table (storage device statistical information)
1020a, 1020b, 1021a, 1021b File system 1040a, 1040b, 1041a, 1041b Volume 1050, 1051 Storage medium group 1060, 1061 Integrated file system

Claims (13)

A storage system comprising a plurality of file management computers that build a file system on a volume of a storage device, and a management computer that manages and redistributes the allocated capacity allocated to each file system,
The management computer is
The file system statistical information related to the allocated capacity for each file system and the storage device statistical information related to the volume capacity of the storage device are collected from the file management computer and the storage device and stored in the storage unit,
When the redistribution request for the allocated capacity of the file system is received, the file system is configured so that the total free capacity of the file system satisfies predetermined performance information based on the file system statistical information and the storage device statistical information. Calculate the allocated capacity for each
Sending a file system reduction instruction to a file management computer that manages a file system whose allocated capacity is reduced among the plurality of file management computers,
Based on the allocated capacity, a volume capacity change instruction is transmitted to the storage device,
A storage system, wherein a file system expansion instruction is transmitted to a file management computer that manages a file system whose allocated capacity is expanded among the plurality of file management computers. The file system statistical information includes an identification ID of the file system, an identification ID of the volume, an allocated capacity of the volume, a free capacity of the volume, and a capacity consumption speed that is a speed at which the capacity is consumed. And
The storage device statistical information includes an identification ID of the volume and an allocated capacity for each identification ID of the volume,
The storage system according to claim 1, wherein the predetermined performance information is the capacity consumption speed. The file system statistical information includes an identification ID of the file system, an identification ID of the volume, an allocated capacity of the volume, an empty capacity of the volume, an access speed determined from a past input / output speed, And the requested access speed required,
The storage device statistical information includes an identification ID of the volume and an allocated capacity for each identification ID of the volume,
The storage system according to claim 1, wherein the predetermined performance information is the requested access speed. The file management computer includes
An integrated file system that integrates multiple file systems is built,
File location management information for associating the integrated file system and the file system with respect to a stored file is stored in the storage unit,
The file management computer
When an input / output request is received for a file of the integrated file system, the file of the input / output request is identified and accessed with reference to the file location management information,
The storage system according to claim 1, wherein the file location management information is updated when a file migrates between the file systems. The storage system further includes:
It has a receiving device that accepts file input / output requests,
The accepting device is:
When an input / output request to a file is received, the file in the file management computer is referred to. When the stub information indicating a link to another file system is stored in the file, the link destination file is stored again. Send an I / O request to the file management computer
The stub information is created when the file moves between the file management computers or within the file management computer, and a link to the destination file is held with the original file name. The described storage system. Allocation assigned to each file system in a storage system and a storage system in which a plurality of file management computers that construct a file system on a volume of the storage device can communicate with the management computer via a network A capacity management method for managing and redistributing capacity,
The management computer includes
File system statistical information related to the allocated capacity for each file system and storage device statistical information related to the volume capacity of the storage device are collected from the file management computer and the storage device and stored in the storage unit,
The management computer is
When the redistribution request for the allocated capacity of the file system is received, the file system is configured so that the total free capacity of the file system satisfies predetermined performance information based on the file system statistical information and the storage device statistical information. Calculate the allocated capacity for each
Sending a file system reduction instruction to a file management computer that manages a file system whose allocated capacity is reduced among the plurality of file management computers,
Based on the allocated capacity, a volume capacity change instruction is transmitted to the storage device,
A capacity management method, comprising: transmitting a file system expansion instruction to a file management computer that manages a file system in which the allocated capacity is expanded among the plurality of file management computers. The file system statistical information includes an identification ID of the file system, an identification ID of the volume, an allocated capacity of the volume, a free capacity of the volume, and a capacity consumption speed that is a speed at which the capacity is consumed. And
The storage device statistical information includes an identification ID of the volume and an allocated capacity for each identification ID of the volume,
The capacity management method according to claim 6, wherein the predetermined performance information is the capacity consumption speed. The file system statistical information includes an identification ID of the file system, an identification ID of the volume, an allocated capacity of the volume, an empty capacity of the volume, an access speed determined from a past input / output speed, And the requested access speed required,
The storage device statistical information includes an identification ID of the volume and an allocated capacity for each identification ID of the volume,
The capacity management method according to claim 6, wherein the predetermined performance information is the requested access speed. The file management computer includes
An integrated file system that integrates multiple file systems is built,
File location management information for associating the integrated file system and the file system with respect to a stored file is stored in the storage unit,
The file management computer
When an input / output request is received for a file of the integrated file system, the file of the input / output request is identified and accessed with reference to the file location management information,
The capacity management method according to claim 6, wherein the file location management information is updated when a file migrates between the file systems. The storage system further includes:
It has a receiving device that accepts file input / output requests,
The accepting device is:
When an input / output request to a file is received, the file in the file management computer is referred to. When the stub information indicating a link to another file system is stored in the file, the link destination file is stored again. Send an I / O request to the file management computer
When the file moves between the file management computers or within the file management computer, the stub information is created and a link to the destination file is held with the original file name. The capacity management method described. In a storage system comprising a storage device and a plurality of file management computers for constructing a file system on a volume of the storage device, a management computer capable of communicating with the storage device and the file management computer via a network There,
The management computer includes a storage unit and a processing unit,
In the storage unit,
A statistical information acquisition program for collecting file system statistical information relating to the allocated capacity for each file system and storage statistical information relating to the volume capacity of the storage device from the file management computer and the storage device;
An allocated capacity calculation program for calculating an allocated capacity for each file system so as to satisfy a predetermined performance information based on the file system statistical information and the storage device statistical information; Remembered,
The processor is
Upon receiving a request to reallocate the allocated capacity of the file system, the file system is reduced to a file management computer that manages a file system whose allocated capacity is reduced among the plurality of file management computers based on the calculated allocated capacity. Send instructions,
Based on the calculated allocated capacity, send a volume capacity change instruction to the storage device,
A management computer that transmits a file system expansion instruction to a file management computer that manages a file system whose allocated capacity is expanded among the plurality of file management computers based on the calculated allocated capacity. The file system statistical information includes an identification ID of the file system, an identification ID of the volume, an allocated capacity of the volume, a free capacity of the volume, and a capacity consumption speed that is a speed at which the capacity is consumed. And
The storage device statistical information includes an identification ID of the volume and an allocated capacity for each identification ID of the volume,
The management computer according to claim 11, wherein the predetermined performance information is the capacity consumption speed. The file system statistical information includes an identification ID of the file system, an identification ID of the volume, an allocated capacity of the volume, an empty capacity of the volume, an access speed determined from a past input / output speed, And the requested access speed required,
The storage device statistical information includes an identification ID of the volume and an allocated capacity for each identification ID of the volume,
The management computer according to claim 11, wherein the predetermined performance information is the requested access speed.

Images